Integrating Fundamental and Applied

Combustion Research

Doug Longman and Stephen T. Pratt

Energy Systems Division

and

Chemical Sciences and Engineering Division

Argonne National Laboratory

Supported by the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, and the Office of Science, Office of

Basic Energy Sciences.

Our Goal

The experimental and theoretical characterization of

chemical reactivity in the context of combustion

Fundamental Science in the Context of Combustion

Novel approaches to quantum chemistry that address current limitations

New methods for systematic thermochemistry

Advanced techniques for generating accurate global potential energy surfaces

New experimental methods for high-temperature chemical kinetics and dynamics

New perspectives on chemical reaction dynamics and kinetics (roaming, intercepting…)

New approaches to global sensitivity analysis and uncertainty analysis

Rational methods for choosing problems and improving combustion models

While the effort is performed in the context of combustion, the work addresses major issues in 21st Century gas-phase chemistry

Connecting Basic and Applied Combustion Research Using Chemistry, Engineering, and HPC to drive 21st Century

Combustion Science and Technology

Coupling large-scale, high-fidelity engine simulations that include detailed fuel

chemistry with global sensitivity analysis allows the identification of key

chemical and physical parameters whose uncertainties determine the

uncertainties in the predictions of the simulations.

Minimizing the uncertainties in these key parameters through new calculations

and experiments can reduce uncertainties in subsequent simulations, and allow

the rational improvement of the models and their predictions.

This approach provides a 21st century, first-principles method to improve

combustion simulations and aid the development of new fuels and engines.

This approach also provides a science-based approach for identifying the most

relevant use-inspired problems for future fundamental studies.

Collaborative work within the CSE, ESD, and ALCF at Argonne has already demonstrated the efficacy of the approach, and we are ready to develop its

potential on a much larger scale.

Active Thermochemical Tables Branko Ruscic

Simultaneous optimization of the full thermochemical network

Consistent propagation of errors

Instantly updated, allowing incorporation of new data

Allows what ifs? and suggests key measurements

A unified approach to the evaluation and compilation of thermochemical data

• • • •

available on the web at: http://atct.anl.gov/

Miniature, High-Pressure, Ultra-Fast Shock Tube

Shock tubes allow the study of high-temperature chemistry but are generally very large (5-10m) and have low repetition rates, making them difficult to use in many environments (e.g., the APS) A unique mini shock tube has been developed, allowing P = 10 – 100 bar; T = 600 – 3500 K; Rep-rate > 4 Hz (vs. 1 – 2 per hour for a normal tube). Reproducible shocks to allow signal averaging! New design allows advanced diagnostics providing new chemically detailed data. -Time-resolved-SAXS at the APS

-Photoionization TOF-MS at ALS

Robert S. Tranter and Patrick T. Lynch

Molecular beam sampling from shock tube

Rational Improvement of Combustion Mechanisms Realistic models require data for thousands of reactions; only a fraction are validated

Michael Davis, Larry Harding, and Stephen Klippenstein

Global sensitivity analysis

identifies key reactions New experiments or theory to

provide improved data Feedback, iterate, and

ultimately automate

C4H9OH + HO2

CH3 + HO2

H2O2 + (M)

Applications to larger systems (butanol)

first sensitivity analysis new rate calculation

Methanol concentration vs. temperature in a high-pressure reactor (Glarborg). - - - is the new model

new predictions and verification

T(K)

[CH

3O

H]

(pp

m)

~5x lower rate

CH3OH + HO2

with D. D. Y. Zhou and R. Skodje (University of Colorado), and A. S. Tomlin (University of Leeds)

Speciation and Sensitivity Analysis in Engine Simulations W. Liu, M. J. Davis, R. Sivaramakrishnan, G. M. Magnotti, P. Kundu, S. Som, and D. E. Longman

Following key elementary reactions under engine conditions:

C7H15O2-2 = C7H14OOH2-4

Temperature contours show where and when these reactions are important

Significant reduction in simulations

necessary for sensitivity analysis

Engine simulations with global sensitivity analysis identify both the important reactions and the conditions under which they must be well-characterized.

Directly couple fundamentals to end use.

Understanding the Role of Non-Boltzmann Reactant

Distributions in Low-Temperature Oxidation

M.P. Burke, C.F. Goldsmith, Y. Georgievskii, and S.J. Klippenstein

Reactions of thermalized and hot reactants may display very different kinetics

Standard approach assumes products are thermalized before undergoing next reaction

In some instances, reactants are still "hot" and the subsequent rates are affected

A method was developed to treat this situation through a series of coupled master equations

The method allows the determination of rate coefficients, k(T,P,XO2), for combustion modeling

Factor of three differences in OH predictions at conditions

commonly used to study low-temperature combustion

Dinner?